In the Big Data ecosystem, the velocity and volume at which data arrive for processing represent two challenging issues to be addressed in the design of systems, frameworks and applications. These challenges are exacerbated by increasingly demanding Quality of Service requirements that must be met despite workload variability or changes occurring in the execution environment, which might leverage on multi–clouds or edge computing resources. Due to the presence of multiple layers that compose a data analytics platform, the variable resource requirements across the layers and the intrinsic complexity of each layer, human–assisted control or manual configuration is unrealistic. Autonomic systems enable to rule the complexity of managing data analytics platforms and integrate monitoring, planning, and execution capabilities so to satisfy some utility goal (e.g., maximize performance, reduce power wastage, guarantee reliable processing). The variety and complexity of Big Data systems, that include data center and cloud resource managers, distributed storage systems, frameworks for batch, micro–batch and data stream processing, demand for specific autonomic solutions to address the multiple facets and foster novel interdisciplinary approaches.<br>This workshop intends to promote a community–wide discussion to identify and find suitable solutions that enable autonomic features in systems, frameworks, and applications for Big Data analytics. We are looking for papers that present new techniques, introduce new methodologies, propose new research directions, or discuss research challenges and report latest efforts from academia and industry that include (but are not limited to) the following topics:<br>– Autonomic provisioning of Big Data applications in Cloud, distributed Cloud and edge computing environments<br>– Autonomic resource management and admission control for Big Data systems<br>– Autonomic data caching, movement and partitioning for Big Data systems<br>– Customizations and extensions of existing software infrastructures and platforms to support autonomic Big data analytics<br>– Elastic techniques to cope with bursty workloads and varying resource demands<br>– Runtime reconfiguration strategies to cope with highly dynamic execution environments<br>– Self–adaptive scheduling and placement strategies for Big Data systems on clusters, Clouds, and distributed Clouds<br>– Applications and case studies of autonomic Big Data analytics in various domains, including astrophysics, biology, climate change, healthcare, Internet of Things, Smart Cities, and social networks<br>
Abbrevation
ASBDA
City
Tucson
Country
United States
Deadline Paper
Start Date
End Date
Abstract